Metabolic biotransformation half-lives in fish: QSAR modeling and consensus analysis

Area of Science:

• Environmental Chemistry
• Toxicology
• Computational Chemistry

Background:

• Bioaccumulation in fish is critical for assessing chemical risk, driven by uptake and elimination rates.
• Hydrophobic organic chemicals pose high bioaccumulation risks, with biotransformation rate constants being key parameters.
• Limited empirical data exists for biotransformation rates, hindering comprehensive hazard and risk assessments.

Purpose of the Study:

• To develop and validate novel Quantitative Structure-Activity Relationships (QSARs) for predicting fish whole-body biotransformation half-lives (HLN).
• To utilize theoretical molecular descriptors capturing whole-molecule characteristics for improved predictive modeling.
• To compare the performance of new QSARs against existing fragment-based methods.

Main Methods:

• Development of three new QSARs using minimal theoretical molecular descriptors (n=9).
• Validation of QSARs employing three distinct data set splitting schemes.
• Comparison with existing QSARs that utilize up to 59 fragment-based descriptors.

Main Results:

• The new QSARs demonstrated strong predictive performance, with external cross-validation Q(2)ext ranging from 0.75 to 0.77 and CCCext from 0.86 to 0.87.
• Prediction accuracy was comparable to existing methods, with RMSE in prediction between 0.56 and 0.58.
• The models provide mechanistic insights and include applicability domain information.

Conclusions:

• The developed QSARs offer a robust and efficient method for predicting fish biotransformation half-lives.
• These models enhance mechanistic understanding and support chemical screening by minimizing false negatives.
• Consensus modeling approaches and identification of data gaps are crucial for prioritizing future research and testing.